AN EFFICIENT APPROACH FOR TEXT MINING USING SIDE INFORMATION

Nowadays, in many text mining applications, information is present in the form of text documents. Text document contains various types of information such as side information or metadata. The different types of information such as document provenance information, title of the document, links in the document, user-access behavior from web logs, or other non-textual attributes treated as side information contained into the text document. Such attributes contains a large amount of information for clustering purposes. It is difficult to estimate the importance of this sideinformation when text document contains some of the information is noisy. In such cases, to avoid the low quality of mining process we need a principled way to perform the text mining, to maximize the advantages from using this side information. Conformation to that, this paper represents solution to the use of side information for clustering by hierarchical algorithm which then extends to the classification problem on real data sets.